Liver stiffness assessed by real-time two-dimensional shear wave elastography predicts hypersplenism in patients with Wilson's disease: a prospective study

Jiajia Wang, Minxia Hu, Qiang Zhu, Lanting SunBmc Medical Imaging 2022 · DOI

TL;DR

This study demonstrates that real-time two-dimensional shear wave elastography (2D-SWE) can reliably predict the development of hypersplenism in patients with Wilson's disease (WD). A liver stiffness cutoff of 10.45 kPa effectively stratifies patients into high- and low-risk groups, enabling earlier intervention and potentially preventing severe complications.

What Problem Does This Paper Address?

Wilson's disease (WD) is a genetic disorder characterized by copper accumulation, often leading to liver damage. A significant complication of WD is hypersplenism, a condition where the spleen becomes overactive, leading to a reduction in blood cell counts. Hypersplenism in WD patients can necessitate risky treatments like splenectomy, which carries its own set of complications, including portal vein thrombosis and overwhelming post-splenectomy infection. Currently, there's a need for non-invasive methods to predict hypersplenism early in WD patients to allow for timely intervention and management. Previous research has explored 4D flow MRI for this purpose, but it has limitations like patient tolerance and cost. This paper investigates whether 2D-SWE, a more accessible ultrasound-based technique, can serve as a reliable predictor of hypersplenism in WD.

Methods

The study employed a prospective longitudinal cohort design. Ninety WD patients with liver involvement but no pre-existing hypersplenism were enrolled.

  • 2D-SWE Measurement: Liver stiffness was assessed using a Mindray-Resona 7 ultrasound system with a specific probe. Measurements were taken from the right liver lobe, ensuring consistency in depth and sampling frame size. The procedure involved patients holding their breath for 5-6 seconds, and measurements were considered reliable when specific motion stability and reliability map indicators were met. The median of five repeated measurements was taken as the final liver elasticity value.
  • Hypersplenism Assessment: Patients were followed up monthly for 24 months or until hypersplenism developed. Hypersplenism was defined by specific leukocyte and platelet count thresholds.
  • Statistical Analysis: Cox regression was used to identify independent risk factors for hypersplenism. Receiver operating characteristic (ROC) curves and area under the curve (AUC) were calculated to assess the predictive performance of liver stiffness, age, and portal vein diameter. A cutoff value for liver stiffness was determined using the maximum Youden index. Kaplan-Meier survival curves and log-rank tests were used to compare hypersplenism incidence between risk groups.

Datasets

The study utilized data from 90 Wilson's disease patients recruited between May 2018 and December 2018 at a single medical center. No new benchmark datasets were introduced.

Key Insights and Ideas

  • 2D-SWE as a Promising Predictor: The study strongly suggests that 2D-SWE is a feasible and effective tool for predicting hypersplenism in WD patients, offering a non-invasive alternative to more complex imaging techniques.
  • Early Stratification for Intervention: The identified cutoff value of 10.45 kPa for liver stiffness allows for the stratification of WD patients into high- and low-risk groups for hypersplenism. This is crucial for guiding clinical management and potentially initiating early interventions like splenic artery embolization.
  • Time-to-Event Difference: The significant difference in the median time to hypersplenism development between high-risk (15 months) and low-risk (22 months) groups highlights the predictive power of liver stiffness in anticipating the timeline of disease progression.
  • Comparison with Existing Modalities: The study positions 2D-SWE as potentially superior to 4D flow MRI for predicting hypersplenism in WD, considering factors like patient tolerance and accessibility.

Key Results

  • Hypersplenism Incidence: 32.2% (29 out of 90) of patients developed hypersplenism during the follow-up period.
  • Independent Risk Factors: Age, portal vein diameter, and liver stiffness were identified as independent predictors of hypersplenism.
  • Predictive Performance: Liver stiffness demonstrated the highest AUC (0.817) for predicting hypersplenism, outperforming age (0.433) and portal vein diameter (0.662).
  • Optimal Cutoff: A liver stiffness cutoff of 10.45 kPa yielded a sensitivity of 75.9% and a specificity of 73.8% for predicting hypersplenism.
  • Risk Group Comparison: Patients with liver stiffness ≥ 10.45 kPa had a significantly higher incidence of hypersplenism (57.9%) compared to those with < 10.45 kPa (13.5%) (p < 0.001). The median time to hypersplenism was also shorter in the high-risk group (15 months vs. 22 months).

Limitations and Future Work

  • Single-Center Study: The study was conducted at a single medical center, which may limit the generalizability of the findings.
  • Single Radiologist: Data was collected by a single experienced radiologist, although the authors note excellent intra-observer reproducibility for 2D-SWE.
  • Limited Lab Indicators: While FIB-4 and APRI were analyzed, the study suggests that including more laboratory indicators might further improve prediction efficacy.
  • Future Directions: The authors emphasize the need for larger, multi-center studies to validate these findings. They also suggest further investigation into the role of 2D-SWE in conjunction with other laboratory markers for enhanced prediction of hypersplenism in WD patients. The potential for 2D-SWE to monitor disease progression and guide treatment adjustments is also highlighted.

Summary updated

BibTeX
@article{wang2022liver,
  author = {Jiajia Wang and Minxia Hu and Qiang Zhu and Lanting Sun},
  title = {{Liver stiffness assessed by real-time two-dimensional shear wave elastography predicts hypersplenism in patients with Wilson's disease: a prospective study}},
  journal = {Bmc Medical Imaging},
  year = {2022},
  doi = {10.1186/s12880-022-00749-x},
}
Figure